Refering to other's ideas

Research indicates that gender differences in performance are related to “common, ordinary differences” in the mathematics and science education of girls and boys (e.g., sex-role stereotyping about mathematics and science skills)––differences that contribute to “different interests, attitudes, achievements, and enrollments during junior and senior high school” (Kahle, 1996, p. 86). These differences can have serious ramifications for girls in terms of their postsecondary education and career choices. High school course-taking, for example, has been shown to have a substantial effect on a student’s postsecondary education. According to Clifford Adelman, of all precollege courses, taking high-level mathematics courses in high school has the strongest impact on a student’s postsecondary education: “Finishing a course beyond the level of Algebra II more than doubles the odds that a student who enters postsecondary education will complete a bachelor’s degree” (Adelman, 1999, executive summary). The Campbell et al. review of the research also showed that a strong high school background, particularly in math, is “key to overall success in college” (Campbell, Jolly, Hoey and Perlman, 2002).

Research also points to several practices that promote an equitable learning environment for girls and have a positive impact on their continuation in quantitative disciplines and science. They include collaborative learning, hands-on experiences, an emphasis on practical applications, and the teaching of science in a more holistic and social context (Campbell et al, 2002; Davis and Rosser, 1996; Hansen et al., 1995; Koch, 2002; Lee, 1997; Wenglinsky, 2000). Many researchers also agree that mentors, role models, and networks are important from the early grades and throughout a woman's career in the sciences (Astin and Sax, 1996; Clewell and Darke, 2000; Hansen et al., 1995; Thom, 2001). Programs for girls combining hands-on activities, role models, mentoring, internships, and career exploration have improved girls' self-confidence and interest in STEM courses and careers and helped reduce sexist attitudes about STEM (Campbell and Steinbrueck, 1996; Ferreira, 2001).

Lastly, in addition to collaborative learning, mentors and role models, Hansen, Walker, and Flom (1995), authors of Growing Smart: What's Working for Girls in Schools, found evidence that girls are more likely to thrive in a learning environment that provides:
· opportunities for leadership and exploration of new ideas;
· active, intelligent engagement with concerned adults and other students;
· consciousness-raising about gender, race, and class issues; and
· single-sex grouping to provide a more supportive context for learning.

These issues and issues of diversity are discussed below.
Learning Styles
Different learning styles based on gender have been suggested as factors related to the STEM participation and achievement of girls and students of color. For example, locus of control has been cited as a factor related to mathematics and science achievement. Some studies have shown that men are more likely than women to attribute success in mathematics to their own ability, while women are more likely to attribute success in mathematics to effort or luck. However, other studies have not found a sex difference in locus-of-control orientation, making the evidence unclear (Clewell and Anderson, 1991). Cognitive style, or the way men and women process knowledge, has also been shown to be related to achievement. Some research suggests that students of color tend to be more “field dependent” (i.e., they process information in a more global than analytic fashion) in their cognitive style than their white peers. A field-dependent cognitive style may inhibit students’ interest and performance in mathematics (Clewell and Anderson, 1991).

Spatial visualization, although more a skill than a learning style or characteristic, is often cited as a factor related to different outcomes in mathematics by gender (Leder and Fennema, 1990). Fennema found that low spatial skills seemed to have a different impact on the achievement of males and females. In one study, males with low spatial skills but high verbal skills had the highest performance on a mathematics achievement test while females with low spatial skills and high verbal skills had the lowest performance. Again, the research is unclear as to whether there are sex differences in spatial visualization ability (Clewell and Anderson, 1991).

In terms of technology, researchers have documented different interactions and attitudes to technology and computers among girls and boys (AAUW, 2000; Kaiser Family Foundation, 1999; Margolis and Fisher, 2002; see also Volman and van Eck, 2001). For example, in a 2000 study conducted by the American Association of University Women (AAUW), girls reported a preference for interacting with people to working on a computer. They were more likely to view the computer as a tool, while boys were more likely to see it as a toy. Girls are more engaged by software that provides opportunities for collaboration, strategizing and critical thinking, and creativity (AAUW, 2000; Kafai, 1998; see also Volman and van Eck, 2001). In contrast, Kafai found that, in designing games, boys were more likely to create adventure games organized around fantasy places. The games boys designed were also more likely to involve violence than those designed by girls (1998). These differences led Davis, et al. to suggest that girls need help seeing the “people benefits” of computer science (and of science and engineering in general) as well as the creative aspects of programming (1996).

Research has also shown that knowledge of computer programming is related to the computing gap between boys and girls. Girls' level of programming skills is a strong predictor of their sense of self-efficacy in computing and of college success in computer science (see Sanders, 2002). Other researchers have found that differences in computer knowledge and skills are closely related to differences in computer experience, again favoring boys (Volman and van Eck, 2001). And, on the question of how to increase girls' participation in computer science, one study found that improving teachers' use of gender-equity strategies did not increase girls' enrollment in advanced placement computer science courses (Sanders, 2002).
Teaching StrategiesStudies suggest that certain teaching strategies may foster the STEM participation and achievement of girls and students of color. For example, some studies have found that cooperative learning groups and active learning motivate young women to study mathematics and science (Bartsch et al., 1998; Ferreira, 2001). Girls' performance in scientific subjects is also enhanced by field trips, labs, and career counseling, which help students see the relevance of science and mathematics in the broader context of work and life (Kahle, 1996). A study of Project Discovery found that participation in classes where teachers were trained in authentic assessment, cooperative learning, grade-appropriate inquiry curricula, and the national standards in mathematics and science significantly decreased the number of boys and girls who thought that “science was for boys” (Kahle and Rogg, 1996).

However, there is also evidence suggesting that cooperative learning may not always increase girls’ participation and achievement. Some research has documented situations where girls were less likely than boys to receive help from boys in cooperative groups, and, if there is only one girl in a group, the boys usually ignore her (Lee, 1997). Jovanovic and King found in their study that boys were more likely to “hog” resources in small, coed settings, while girls were more often passive participants (1998). Volman and van Eck's review of research on information technology also found evidence that boys tend to dominate computer-related activities (2001). Another study found no achievement gains for boys or girls associated with increased cooperative group work in high school biology classes (Kahle, 1996, cited in Boone and Kahle, 1998).

In their review of literature, Hansen, Walker, and Flom found that hands-on experiences such as handling tools and equipment may boost girls' interest in mathematics and science (Hansen, et al, 1995). Another study found that hands-on engineering activities made girls six times more likely to consider engineering as a career (Campbell and Shackford, 1990). However, contrary to their expectation, Jovonovic and Dreves found that hands-on science classrooms did not reduce the gap between boys' and girls' science attitudes (1998). They concluded that “boys and girls experience hands-on science classrooms differently.” Similar to the findings on cooperative learning, both Hansen et al. (1995) and Jovonovic and Dreves (1998) found that boys tend to dominate science-oriented activities, especially those involving special equipment.

There is some evidence that using scientific equipment and hands-on activities are related to higher science and mathematics achievement. The Campbell, et al. (2002) report, Upping the Numbers, points to several studies providing such evidence, including the following:
· National Assessment of Educational Progress (NAEP) science achievement scores were higher for nine-year-olds who used equipment like meter sticks, scales, and compasses in class (Campbell, Hombo, Shackford and Mazzeo, 2000).
· NAEP mathematics scores were higher for eighth graders who participated in hands-on learning activities than for those who did not. They were also higher for 17-year-olds with access to computers to learn mathematics and solve mathematical problems (Wenglinsky, 2000).
· Participating in physical science laboratory activities improved girls' achievement, while not affecting that of boys (Burkham, Lee and Smerdon, 1997; Lee, 1997).
Learning Environments
Closely related to teaching strategies, certain educational environments may help increase the STEM participation and achievement of girls and students of color.

Related to the findings about the impact of mentors and role models, studies have found that support from adults can play a key role in encouraging girls. One Girls Incorporated study, The Explorer's Pass, showed that girls in mathematics and science classes and programs benefited from adult encouragement and modeling to overcome “a reluctance to get dirty and a tendency to ask for adult rescue when a task seemed difficult or boring” (Girls Incorporated, 1991, p. viii). The study also found that girls needed a supportive environment to pursue interests, take (reasonable) risks, not fear making mistakes, and use “mistake making” as a method of learning. Another AAUW report, Girls in the Middle: Working to Succeed in School, showed the importance of adults fostering an atmosphere of respect for girls’ voices and approaches to learning, whether or not they conform to the dominant culture of the school (Cohen et al., 1996).

Although several scholars support single-sex grouping as a way to provide a supportive learning environment for girls, the research around the long-term effects of single-sex education are inconclusive (Davis et al., 1996; Phillips, 1998). AAUW's review of research on this issue determined that single-sex learning environments in primary and secondary schools do not necessarily eliminate sexism or lead to increases in achievement for girls (1998). In fact, some research has shown single-sex environments to be more sexist than coeducational environments (Lee, 1997). Research to date has also not supported long-term gender segregation in mathematics and science classes as strengthening girls' interest, achievement, and persistence in STEM fields (Leder, 1990, cited in Davis, 1996). Many researchers and educational reformers fear that single-sex learning environments are too simplistic a way to address the complex issues related to providing equitable and supportive learning environments for girls and may detract from efforts to make coed schools more equitable (Bailey, 1996; Campbell and Wahl, 1998a; Campbell and Wahl, 1998b).
Issues of Diversity
Women of color in the sciences face the double barriers of racism and sexism. However, little research has explored the relationship of gender and ethnicity in terms of girls' STEM achievement. Very limited data are available disaggregated by sex and racial/ethnic group (Coley, 2001; Kahle, 1996); more often, data are presented by either sex or race/ethnicity. One exception is a recent Educational Testing Service report, which provides some data, including NAEP, SAT, high school course-taking, advanced placement, educational attainment, and employment data by gender and race/ethnicity (Coley, 2001). Coley found that across these indicators, gender differences did not vary much by race and ethnicity. Females outperformed males on some indicators while males outperformed females on others. He concluded that the “nature of gender inequality in education is a complex phenomenon,” and noted that both gender and race/ethnicity are “crucial factors” that must be attended to (see attachment for a summary of achievement data results by gender and race/ethnicity).

Clewell and Ginorio found in their review that research on Caucasian women and girls is not generalizable to women and girls of color. Neither is research on women and girls of one race/ethnicity or social class generalizable to other race/ethnicities or social classes (Clewell and Ginorio, 1996). The majority of research on girls and women of color has been conducted with African-American girls and women, followed by Latinas. Three studies found that at the time of school entry, race, social class, and gender differences in mathematics readiness were small. One study found that in the early grades, the link between parental and child expectations (e.g., that boys will perform better in mathematics) was weaker in schools with students from low-income families than in schools with middle-class students (Clewell and Ginorio, 1996).

While research has shown that some learning styles influence different achievement in mathematics and science for women and people of color, there is no research on the learning styles of girls of color and the ways in which sex and race or ethnicity interact to influence learning (Clewell and Ginorio, 1996). NAEP data showed that girls’ experiences in mathematics and science differed by race/ethnicity. White students had more science experiences than African-American students, and the difference increased with age. It also showed that African-American girls at ages nine and 13 have conducted the fewest science experiments of all racial/ethnic groups (Clewell and Ginorio, 1996).

In terms of persons with disabilities, data are even more limited (Bauer, 2001; National Science Foundation, 2000). Available data indicate that girls with disabilities are among those least likely to have mathematics and science experiences (Wahl, 2001). Students with disabilities of either gender are unlikely to take advanced course work in mathematics and science, and few disabled students pursue higher education. Further, there are substantial differences between disability groups, which necessitates the disaggregation of data by type of disability. For example, achievement outcomes are very different for students with visual impairments compared with those with physical or mental disabilities (Wahl, 2001).

Characteristics of Effective STEM and Afterschool Programs
An estimated three to four million (20% to 25%) low- and moderate-income urban children participate in afterschool programs, and the number appears to be growing (Halpern, 2002). Attention to afterschool hours has increased substantially in the last decade as policymakers, child development professionals, and parents see this time period as “one of unusual risk and opportunity” (Hofferth, 1995, cited in Halpern, 2002). The risks range from boredom to self-destructive behavior and crime on the part of young people, while the opportunities for youth include developing caring relationships with peers and adults and taking part in academic enrichment and support (Halpern, 2002). Halpern suggests that, after home and school, afterschool programs are becoming a “third critical developmental setting for low- and moderate-income children.”

One nationally representative study of afterschool programs indicated that in 1991, 1.7 million kindergarten through eighth-grade children were enrolled in approximately 50,000 programs (Seppanen et al., 1993). This study found tremendous variability in program characteristics such as sponsorship and location. It is not clear how many children and youth are enrolled in afterschool programs that focus on science, technology, engineering, and mathematics. However, as noted earlier, studies have documented that girls have fewer computer and science-related experiences outside of school than boys (Farenga, 1995; Kahle et al., 1993; see also Volman and van Eck, 2001).

Studies have also shown that afterschool participation contributes to reduced drug use and juvenile crime, and lower dropout and teen pregnancy rates among youth, as well as higher standardized test scores and college attendance rates, better handling of conflicts, more cooperative relationships, better social skills, and improved self-confidence (Fashola, 1998; Holloway, 1999; Huang, 2000; Afterschool Alliance, 2001). Specifically, Fashola concluded, based on an extensive review of programs, that the following qualities were related to increased student academic achievement: greater programmatic structure (e.g., scheduled activities, planned curriculum); a strong link to the school-day curriculum; well-qualified and well-trained staff; and opportunities for one-to-one tutoring (Fashola, 1998). The U.S. Departments of Education and Justice (2000) found that characteristics of high-quality afterschool programs of any type (not just those intended to increase academic achievement) include clear program goals, strong leadership and effective managers, skilled and qualified staff, ongoing professional development, and low adult-to-child ratios (U.S. Departments of Education and of Justice, 2000).
A Youth-Development Framework
Several researchers note that effective afterschool programs are not simply replications of the regular school-day curricula. Children who are not successful in school are not likely to be any more successful in an afterschool program that provides “more of the same” (Scarf and Woodlief, 2000). Recent research suggests that programs adhering to a youth-development framework are more likely to promote positive youth outcomes (James and Jurich, 1999; McLaughlin, 2000; National Research Council and Institute of Medicine, 2002; Roth and Brooks-Gunn, 1998). Research related to effective youth-development programs are summarized below.

In a review of research related to community-based programs, Eccles and Gootman found that characteristics linked to promoting positive development and outcomes in adolescents include safety and security, a structure that recognizes adolescents' increasing social maturity, and strong links between families, schools, and community resources. Characteristics of effective programming and a youth-development framework also included opportunities for youth to:
· experience supportive relationships and receive emotional and moral support;
· feel a sense of belonging;
· be exposed to positive morals, values, and positive social norms;
· be efficacious, to do things that make a real difference, and play an active role in the program; and
· develop academic and social skills, including learning how to form close relationships with peers that support and reinforce healthy behaviors, as well as acquire the skills necessary for school success and a successful transition to adulthood (National Research Council and Institute of Medicine, 2002, p. 117).
In four compendia of evaluations of effective youth programs, the American Youth Policy Forum (AYPF) presented summaries of 199 youth program evaluations. The analysis of these programs yielded a list of characteristics similar to that of Eccles and Gootman. Specifically, the forum listed the following characteristics as contributing to program effectiveness:

· high quality implementation;
· high standards and expectations of youth;
· parent/guardian participation and community involvement;
· viewing youth as resources;
· provision of long-term services, supports and follow-up;
· caring, knowledgeable adults; and
· community service, service-learning, work-based learning (American Youth Policy Forum, 1997,1999, 2000, 2001).

McLaughlin's research on community-based organizations (CBOs) also supports these qualities as relevant in terms of program effectiveness. Specifically, organizations that are youth-, learning, and assessment-focused had a significant impact on the skills, attitudes, and experiences of the at-risk youth they served (McLaughlin, 2000). For example, youth who participated in CBOs characterized by these qualities had higher academic aspirations, greater self-confidence and optimism, and stronger feelings of civic responsibility compared with American youth generally.

Roth and Brooks-Gunn (1998) found that high-quality outcome evaluations of youth-development programs are scarce, but a review of the literature showed that programs incorporating more elements of youth-development approaches showed more positive results. They also found that longer-term programs engaging youth throughout adolescence appeared most effective. However, Roth and Brooks-Gunn noted that, while there was some evidence of the effectiveness of the youth-development framework, many questions remained. For example, it was not clear what mix of youth-development program characteristics produced what outcomes and for whom.
Gender Issues and Afterschool Programs
Scarf and Woodlief (2000) reported that much of the literature on afterschool programs deals with diversity issues on a very general level, typically relating to issues of cultural sensitivity. Very little is written about specific gender issues in youth-development and afterschool programs, and there are few guidelines and recommendations for practice and implementation of gender-equitable programs other than to avoid stereotyping activities by gender. In fact, even in teacher education texts, gender is barely mentioned. Zittleman and Sadker (2002) found in their content analysis of methods texts that, on average, only slightly more than one percent of content dealt with gender issues. Even more limited are guidelines and recommendations for afterschool programs serving girls with disabilities (Froschl, Rousso, and Rubin, 2001).